Joint Bone Spine 80 (2013) 70–76
Available online at
Impairment of two types of circulating endothelial progenitor cells in patients
with glucocorticoid-induced avascular osteonecrosis of the femoral head
Chao Chena, Shuhua Yanga,∗, Yong Fenga, Xinghuo Wua, Dong Chena, Qian Yua, Xiaohong Wanga,
Jing Lib, Juan Chenc
aDepartment of Orthopedics, Union Hospital, Tongji Medical College, Science and Technology of Huazhong University, Wuhan 430022, China
bDepartment of Integrated Traditional and Western Medicine, Tongji Hospital, Tongji Medical College, Science and Technology of Huazhong University, Wuhan 430030, China
cDepartment of Clinical Oncology, The University of Hong Kong, Pokfulam, Hong Kong, China
a r t i c l ei n f o
Accepted 22 February 2012
Available online 13 April 2012
Endothelial progenitor cells
Endothelial colony forming cells
a b s t r a c t
of the femoral head (ANFH).
Methods: Early EPCs and endothelial colony forming cells (ECFCs) were obtained from 33 patients with
glucocorticoid-induced ANFH and 33 age- and sex-matched control subjects. Cells were isolated, in vitro
cultured and studied by Flow Cytometry and Immunofluorescence. Colony-forming unit counts were
observed from 33 patients and 33 healthy controls. Growth kinetics, migratory capacity to multiple
chemo-attractants, in vitro tube formation capacity and cytokine (vascular endothelial growth factor
and stromal cell-derived factor-1) levels in supernatants of two types of EPCs were assayed in ANFH
patients and matched controls (n=4).
Results: Mean numbers of colonies formed by both types of EPCs were decreased in ANFH patients (Early
EPCs: 2.42±1.46 versus 4.52±2.00, p<0.05; ECFCs: 0.62±0.55 versus 1.12±0.82, p<0.05,). Early EPCs
from ANFH patients showed impaired migratory capacity (63.8±11.7 versus 152.3±12.4, p<0.001) and
VEGF secretion (50.8±7.2pg/ml versus 62.8±10.1pg/ml, p<0.05). ECFCs from ANFH patients showed
decreased tube formation capacity (7.1±2.7 versus 23.8±4.3, p<0.001) and proliferation.
Discussion: Early EPCs and ECFCs were impaired in number and function in GC-induced ANFH, and their
distinct reduced capacity profiles might reflect different roles they played in endothelial dysfunction of
© 2012 Published by Elsevier Masson SAS on behalf of the Société Française de Rhumatologie.
Practically avascular osteonecrosis of the femoral head (ANFH),
which represents a devastating condition with a poor prognosis, is
a debilitating skeletal disorder affecting especially young patients
in their third to fifth decades of life. Avascular osteonecrosis, oth-
erwise known as aseptic necrosis of bone, results from multiple
pathological conditions including ischemia, osteocyte apoptosis
and necrosis, impaired vascular repair and subsequent loss of
structural integrity of the articular surface, thus finally leading to
significant clinical morbidity . The most common cause of ANFH
is trauma, and the main cause of non-traumatic ANFH is the use of
glucocorticoid (GC). Although numerous investigators had postu-
lated several mechanisms involved in ANFH, the exact etiology and
∗Corresponding author. Tel.: +86 02 78 53 51 627; fax: +86 02 78 57 18 159.
E-mail address: email@example.com (S. Yang).
pathophysiology of GC-induced non-traumatic ANFH still remains
Recently, growing evidences indicate that microcirculatory dis-
ANFH . Regional endothelial dysfunction due to continuous
exposure to GC in femoral heads results in various disorders such
as aberrant vasoactive substances levels, decreased blood flow,
endothelial cells apoptosis, microcirculatory thrombus formation,
functional microvessels rarefaction, inhibition of angiogenesis and
finally osteoblasts and osteocytes death . Therefore, this has
prompted investigations on the role of endothelial cells and more
importantly, its progenitors – endothelial progenitor cells (EPCs) in
EPCs initially described by Asahara et al.  are now being
broadly investigated because of its significant role in vascular
different disease states and cardiovascular risk [5,6]. Thus, a num-
ber of studies have already implied EPCs as a surrogate biomarker
1297-319X/$ – see front matter © 2012 Published by Elsevier Masson SAS on behalf of the Société Française de Rhumatologie.
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
in vascular dysfunction . Our previous study also had found the
relationship between EPCs and ANFH .
However, recent studies indicated that there was another
newly so-called endothelial colony forming cells (ECFCs) or late
outgrowth endothelial progenitor cells (OECs) demonstrate highly
proliferative potential, exhibit self-renewal capacity and are capa-
ble of de novo vessel formation in vivo. In contrast, the early EPCs
(also known as colony-forming unit endothelial cells or CFU-ECs)
expressing hematopoietic markers lack the proliferative capacity
characteristic of true progenitor cells, fail to form vascular net-
are actually hematopoietic-derived monocytes manifesting some
angiogenic properties . And it seems that ECFCs may serve as a
potential cell source more effectively for cellular therapies aiming
to enhance neovascularization in ischemic area.
Taken together, there is a lack of detailed data concerning the
tioned whether alterations of ECFCs existed in GC-induced ANFH.
Considering significant debate remains about the identity of EPCs,
we conducted a study to investigate the levels of early EPCs and
ECFCs and compared their functions in GC-induced ANFH patients
and age- and sex-matched control subjects.
2.1. Study population
The study population consisted of 33 patients (24 men, 9
women; mean age 42.7 years, ranging from 24-53 years) with
GC-induced ANFH, who were recruited in our hospital (Union Hos-
pital, Wuhan, China) from July 2008 to December 2010, and 33
age- and sex-matched control subjects, who were healthy volun-
teers. Clinical characteristics for all participants were summarized
in Table S1, Supplementary data. ANFH was diagnosed by radiog-
raphy and magnetic resonance imaging (Steinberg stage I to stage
IV). Exclusion criteria were cancer, pregnancy, diabetes mellitus,
impaired renal function and presence of cardiovascular diseases.
Patients with a demonstrable history of direct injury of the hip
were also excluded. This study was approved by the local ethics
committee and written informed consent was obtained from all
2.2. Preparation of mononuclear cells
Peripheral blood sample (60mL) was collected in heparin
sodium tubes from each patient. After density gradient centrifu-
gation, Buffy coat mononuclear cells (MNCs) were collected and
washed three times in PBS with 2% fetal bovine serum (Hyclone,
Logan, UT). MNCs were re-suspended in EGM-2 medium (Lonza,
San Diego, CA), half of which were used to culture ECFCs and the
other to culture early EPCs.
2.3. Culture of early EPCs
MNCs were seeded on tissue culture plates precoated with
fibronectin (BD Biosciences, Bedford, MA) in EGM-2 medium. After
collected and replated at 1×106cells/well onto fibronectin-coated
24-well plates. Growth medium was changed every three days.
Early EPCs colonies were identified as elongated sprouting cells
radiating from a central cluster of round cells and enumerated
2.4. Culture of ECFCs
ECFCs were cultured according to the protocol of Ingram DA
et al. . ECFCs colonies that appeared between 7 and 30
days of culture were identified as well-circumscribed monolay-
ers of cobblestone-appearing cells and enumerated blindly by two
independent investigators on day 14 by visual inspection under
microscope. Subconfluent cells were trypsinized and replated onto
75-cm2tissue culture flasks coated with type I rat tail collagen (BD
Biosciences) for further passage.
2.5. Flow cytometry analysis
and resuspended in PBS. Then 0.5-1.0×106cells were incu-
bated with conjugated antibodies CD133-Allophycocyanin (APC)
(eBioscience, San Diego, CA) and VEGFR2-Peridinin chlorophyll
protein-Cy5.5 (PerCP-Cy5.5) (Biolegend, San Diego, CA) for 20min
at 4◦C away from light. Isotype-identical antibody served as
negative controls. Cells were analyzed using a Beckman Coulter
FC500 flow cytometer equipped with the Cellquest software pro-
gram. Light scatter profiles were collected on the cells using two
electronic gates to discriminate lymphocytes and monocytes in
accordance with previous reports . Results were expressed as
percent of double positive cells.
2.6. Immunofluoresence of ECFCs
ECFCs at early passages (p2 or p3) were plated on type I rat
tail collagen. When reaching 80–90% confluency, cells were fixed
in 2% paraformaldehyde for 20min, followed by permeabilization
with 0.1% Triton X-100. After that, cells were blocked in 10% nor-
mal goat serum and then incubated with primary antibodies at
4◦C overnight. Primary antibodies were diluted as follows: rabbit
antihuman CD31 1:100 (Abcom, Cambridge, UK), rabbit antihu-
man VWF (Von Willebrand Factor) 1:200 (Abcom). Subsequently,
the cells were incubated with goat antirabbit secondary antibody
conjugated to fluorescent isothiocyanate (FITC) and then in 4?,6-
diamidino-2- phenylindole dihydrochloride (DAPI, sigma, St. Louis,
MO) for nuclear staining. Images were obtained using an Olympus
FV500 confocal laser-scanning microscope.
2.7. Uptake of DiI-Ac-LDL and binding of UEA-1 lectin
Both early EPCs and ECFCs were incubated with 2.4?g/mL
Dil-Ac-LDL (Introvogen-Molecular Probes, Eugene, OR) in EGM-2
medium at 37◦C for 4h. Then cells were washed with PBS, fixed in
4% paraformaldehyde, and further incubated with 10?g/mL FITC-
slides were washed with PBS before visualization under confocal
laser scanning microscope.
2.8. Growth kinetics of early EPCs and ECFCs
and four age- and sex-matched controls were enumerated at the
first passage using a trypan blue assay. Then cells were replated
to six-well plates at 1×105/well in EGM-2 and allowed to attach
hemocytometer every three days. Each experiment was performed
in quadruplicated. To make a growth kinetics curve, population
doubling time (PDT) were determined according to the equation:
PDT=T×lg2/lg (Nt/N0), where T was the time interval between
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
each passage, N0was the number of cells seeded and Ntwas the
number of cells at harvest.
2.9. In vitro tube formation assay
Matrigel (BD Biosciences) was used to assess vascular tube
formation according to the manufacture’s instructions. In brief,
1×104/well ECFCs at passage 2–3 or early EPCs at day 7 were
placed on top of Matrigel in 96-well plate and incubated at 3◦C for
48h. Cells were observed every 4h under microscope. The number
of closed network units formed by cells in 4 wells was manually
2.10. Migration assay
Early EPCs and ECFCs were detached and resuspended in EBM-
2, and 3×104cells in 300?L medium were seeded on the top of a
Transwell insert (Corning Costar, Lowell, MA) and then placed over
500?L complete EGM-2 to create a positive chemotactic gradient
in a 24-well culture plate. After 24h incubation, cells on the bot-
tom of the insert were fixed in paraformaldehyde and stain with
0.1% crystal violet (Sigma) and visualized by light microscopy. Cells
microscopic fields using an image analyzing software, Image-Pro
2.11. Measurement of cytokine concentration of supernatant
To evaluate the secretion capacities of EPCs, we measured the
concentration of VEGF and SDF-1 in the supernatant of culture
in EBM-2 medium for 3 days, and then the supernatant was har-
vested. Commercially available ELISA kit (Quantikine for VEGF and
SDF-1, R&D Systems, Minneapolis, MN) was used. Concentration
was determined by comparison with a standard curve according to
the manufacturer’s instructions. Each experiment was performed
3. Statistical analysis
tion and are reported with a 95% confidence interval, unless
otherwise specified. Comparisons of the continuous normally dis-
tributed variables were performed by paired Student’s t-tests
(two-tailed). Categorical variables were compared using Chi2tests
(Fisher’s exact test). Analysis were performed using the statistical
and p<0.05 was considered significant.
4.1. Clinical characteristics of study subjects
We compared clinical characteristics of ANFH patients and
control subjects in Table S1, Supplementary data. Factors (smok-
ing, alcoholism, hypertension, and hypercholesterolemia) that are
thought to be correlated with the number and function of cir-
culating EPCs were evaluated. And factors (use of glucocorticoid,
thrombophilia) involved in GC-induced ANFH were also deter-
mined. The baseline characteristics were similar except that Total
cholesterol, HDL cholesterol and Triglycerides were significantly
elevated (p<0.001, p=0.048 and p<0.001, respectively) in ANFH
Fig. 1. Cell culture and enumeration of early EPCs and ECFCs colonies from adult
peripheral blood. A. Representative photomicrograph (400×) of an early EPCs
colony characterized by a central cluster of rounded cells surrounded by radiating
thin, flat cells. B. Representative photomicrograph (400×) of ECFCs stained with
crystal violet. C. ECFC colony characterized by cobblestone-like morphology (40×).
Number of early EPCs colonies (D) and number of ECFCs colonies (E) per well in
ANFH were decreased when compared with age- and sex-matched healthy controls
(HC). Results represent the mean values±standard derivation of 33 independent
subjects. *p<0.05 by paired Student’s t-test (two-tailed).
4.2. Reduced two types of EPCs colony counts in ANFH
Using a microscope, two types of EPCs colonies were observed
in culture from the mononuclear cells of control groups and
ANFH groups. Early EPCs colonies developed after one week, and
were characterized by a central cluster of round cells surrounded
by radiating thin flat cells (Fig. 1A). ECFCs colonies appeared
after two weeks and varied in size, but all formed a typical
cobblestone monolayer (Fig. 1B, C). Interestingly, we established
that both the number of early EPCs colonies and ECFCs colonies
were decreased in ANFH patients when compared with age-
and sex-matched control subjects (2.42±1.46 colonies/well ver-
sus 4.52±2.00 colonies/well, p<0.05 and 0.62±0.55 colonies/well
versus 1.12±0.82 colonies/well, p<0.05, respectively) (Fig. 1D, E).
Both early EPCs and ECFCs in two groups were stained positively
for DiI-Ac-LDL and FITC-UEA-1 by confocal immunofluorescence
microscope confirming their endothelial phenotype (Fig. 2A–F). In
cific markers of endothelial cells, while the former made up a large
portion of endothelial cell intercellular junctions and the latter was
synthesized exclusively by endothelial cells and functioned in the
maintenance of homeostasis (Fig. 2G–L).
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
Fig. 2. Representative photographs of immunofluorescence staining of early EPCs
and ECFCs. Early EPCs stained by UEA-1 lectin (green) (A), Dil-Ac-LDL (red) (B), and
merged (C) in confocal laser scanning microscope. Bar=10?m. ECFCs stained by
UEA-1 lectin (D), Dil-Ac-LDL (E), and merged (F) by confocal laser scanning micro-
(G, J). Bar=20?m.
4.3. CD133+VEGFR2+cells are not reduced in ANFH
When we analyzed early EPCs on day 7 from ANFH (n=20)
and HC (n=20) by Flow cytometry using antibodies to CD133
and VEGFR2, there was no statistical significant difference in the
percentage of CD133+VEGFR2+ cells in two groups (1.81±0.37%
versus 1.96±0.33%, p=0.092) (Fig. 3A–D).
4.4. Growth kinetics of early EPCs and ECFCs in ANFH
In a comparison of the proliferative kinetics between ANFH and
control groups, we counted the expanded early EPCs and ECFCs
from four different cultures every 3 days. We observed that early
exhibited decreased proliferation in ANFH patients compared to
control subjects (Fig. 4A). PDT also showed a significant difference
in ECFCs growth between groups (6.45±0.40 versus 4.22±0.24,
p<0.05) (Fig. 4B).
4.5. Impaired tube formation of ECFCs in ANFH
To assess the capacity of vasculogenesis, tube formation assay
was performed on both early EPCs and ECFCs using a Matrigel
model. Results showed marked difference. Early EPCs from both
day 7. A. Position and identification of lymphocyte gate (P1) by light scatter pattern.
(B) Subsequently gates were drawn around the high-expressing distinct population
of CD133+cells, and (C) double-positive events for CD133+and VEGFR2+(CD309)
were determined (Q2-2). The percent of double positive cells were calculated on the
basis of the lymphocyte counts. D. Percentage of CD133+VEGFR2+cells of early EPC
t-test (two-tailed). Data are represented as mean values±standard derivation.
D1 D3D6 D9 D12 D15 D18 D21 D24 D27 D30
Number of cells (105)
Fig. 4. Growth kinetics of early EPCs and ECFCs. A. Representative growth curves of
early EPCs and ECFCs from ANFH and healthy controls (HC) during 30 days culture.
B. Population doubling time (PDT) of ECFCs was significantly decreased in ANFH.
Results represent the average of 4 independent experiments. Data are represented
as mean values±standard derivation. *p<0.05 by paired Student’s t-test.
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
and became spindle-shaped but failed to form tube-like structures
(Fig. 5A, B). Of interest, ECFCs organized into capillary structures
with enclosed area successfully, but were impaired in potential to
augment angiogenesis in vitro in ANFH patients (Fig. 5C, D). The
number of closed network units in ANFH was significantly lower
4.6. Impaired migratory function of early EPCs in ANFH
In migration assays, we found that early EPCs from ANFH
patients showed impaired migratory functions (63.8±11.7 versus
152.3±12.4, p<0.001) (Fig. 5E, F, J). However, there were no sig-
nificant differences in migratory functions of ECFCs between two
groups (248.1±19.8 versus 248.9±20.0, p>0.05) (Fig. 5G, H, K).
4.7. Decreased VEGF levels of early EPCs in ANFH
Vascular endothelial growth factor (VEGF) is one of the most
common angiogenic cytokines and stromal cell-derived factor-1
in ANFH compared to control subjects (50.8±7.2pg/ml versus
62.8±10.1pg/ml, p=0.001) (Fig. 6A). But VEGF levels secreted by
ECFCs in two groups showed no difference (32.8±10.5pg/ml ver-
sus 34.0±14.2pg/ml, p=0.554). And SDF-1 levels secreted by both
versus 226.9±27.2pg/ml, p=0.438 and 221.5±36.4pg/ml versus
229.7±39.1pg/ml, p=0.318, respectively) (Fig. 6B).
One theory of the pathogenesis of GC-induced ANFH suggests
that there is a basic defect of microcirculatory balance in this con-
dition, resulting in chronic regional ischemia and endothelial cell
impairment [3,13,14]. In our previous steroid-induced osteonecro-
sis rabbit models, cytokine-induced mobilization of bone marrow
stem cells increased new vessel formation and thus promote func-
tional bone repair of early stage ANFH [15,16].
In this study, we analyzed the feasibility of isolating both early
EPCs and ECFCs from the peripheral blood of GC-induced ANFH
patients and control subjects, and then we compared the functions
of two types of cells. We reported that both early EPCs and ECFCs
were impaired in number and function in GC-induced ANFH but
they showed different reduced capacity profiles. First, we demon-
strated decreased circulating early EPCs and ECFCs numbers in
GC-induced ANFH than age- and sex-matched control subjects as
evidenced by the appearance of EPCs colonies in culture. In addi-
tion, we found most early EPCs did not proliferate and eventually
died out, which was in agreement with previous reports , and
induced ANFH. In tube formation assays, early EPCs failed to form
tubes, which was in line with other studies , whereas ECFCs
made tubes successfully but the ability was also impaired in GC-
induced ANFH. When comparing migration capacity, early EPCs
from ANFH patients showed reduced migratory function but ECFCs
seemed not affected. Furthermore, early EPCs secreted more VEGF
than ECFCs, but the secretory function was impaired in GC-induced
ANFH. Based on these results, we interpreted that the decreased
lial dysfunction in GC-induced ANFH and their distinct reduced
capacity profiles might reflect different roles they played. This was
consistent with several previous studies which suggested the cir-
culating EPC number was decreased in some clinical conditions,
such as hypercholesterolemia, hypertension, diabetes mellitus,
Fig. 5. Functions of early EPCs and ECFCs from four ANFH patients and four age- and
sex-matched control subjects. Early EPCs from control subjects (A, ×40) and ANFH
patients (B, ×40) failed to form tube-like structures on Matrigel. ECFCs made tube
formation successfully in control subjects (C, ×40) and in ANFH patients (D, ×40).
Early EPCs from control subjects (E, ×200) and ANFH patients (F, ×200) showed
migratory capacity to cytokines. ECFCs from control subjects (G, ×200) and ANFH
patients (H, ×200) also showed migratory capacity to cytokines. The functional
capacity for tube formation of ECFCs was reduced in ANFH (I). Early EPCs from ANFH
showed no difference in migratory capacity (K). HPF=high power field. Data are
shown as mean values±standard derivation. P values were determined by paired
Student’s t-test. *p<0.001.
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
SDF-1 levels secreted by early EPCs and ECFCs showed no difference between ANFH
experiments. P-values were determined by paired Student’s t-test.
limb ischemia, rheumatoid arthritis, atherosclerosis, endogenous
endothelial repair and cardiovascular diseases [4–7,18].
Although both early EPCs and ECFCs can be consistently iso-
lated by different methods, attempts to identify the phenotype
of these cells by FACS have proven to be difficult. Numerous
surface antigens have been associated with EPCs: CD34, CD133,
VEGFR-2, CD45, CD14, CD31, CD105, CD144, CD146, von Wille-
brand factor (vWF), c-Kit, Tie-2, E-selectin, and others [4,18–20].
In our present study, when using a combination of CD133 and
VEGFR-2, we did not find a significant difference in the number
of CD133+VEGFR2+cells identified by FACS between GC-induced
ANFH and control subjects. However, we found that the number of
early EPCs colonies and ECFCs colonies were reduced in ANFH. This
circulating cells defined as either early EPCs or ECFCs. Actually, the
precise surface antigen profile of EPCs has not yet fully understood
Despite much controversy concerning the definition and iden-
tification of EPCs circulating in peripheral blood, ex vivo expansion
of these cells is widely used in a wide array of experimen-
tal studies and clinical trials. To date, there are two distinct
EPC phenotypes being identified based on in vitro characteris-
tics. The “endothelial colony-forming cells” (ECFCs), which derived
from CD34+CD45−population display a clonal phenotype and
exhibit essential progenitor characteristics, having high prolifer-
ation potential, self-renewal capacity, and de novo blood vessel
formation in vivo [11,21,22]. Unlike highly proliferative ECFCs,
early EPCs displaying endothelial-like markers and behavior have
a monocyte/macrophage phenotype better described as an angio-
genic macrophage. However, both cell types may participate in
normal vasculogenesis and pathological repair processes [23–25].
Circulating EPCs constitute a circulating pool of cells that sup-
port the re-endothelization at sites of endothelial injury, thus
contributing directly to the homeostasis and viability of the vas-
culature . They home to sites of intima damage following
gradients of cytokines released by injured endothelium and tis-
sues, where they can actively repair the endothelial layer and lead
to reperfusion of ischemic regions within the tissue . The exact
mechanisms on the close relationship of EPCs abnormalities in GC-
induced ANFH remain to be determined. It is reported that GC has
abundant effects on endothelial cells that line the sinusoids and
innerlayer of blood vessels in the femoral head . Long-term
exposure to GC directly leads to endothelial injury and apoptotic
tor cells might also be injured and sensitive to apoptosis induction,
as we found decreased EPCs colonies in GC-induced ANFH. Another
mechanism appears to be related to decrease fibrinolytic activ-
ity, accompanied by lack of mitogens and angiogenic factors that
stimulate endothelial migration and proliferation, which may be
a consequence of increased PAI-1 . This is in accordance with
our study that the proliferation and migration potential of EPCs
were significantly impaired. Moreover, GC can decrease the syn-
thesis of VEGF , which act directly on endothelial cells and
induce angiogenesis during repair process, as observed in our
study that VEGF levels secreted by early EPCs were significantly
decreased in ANFH. In addition, GC excess increased oxidative
stress and thereby perturbs nitric oxide (NO) availability in the
vascular endothelium, which is produced by EPCs themselves and
maintain the optimal environment to promote their mobilization
and expansion . Therefore, it may be assumed that reduced
EPCs number and function implies the reduction of NO availability
and may reflect vascular endothelial dysfunction, thus a potential
mechanism for glucocorticoid-induced ANFH. However, these pos-
sibilities need further investigation to clarify how GC impact on
EPCs in glucocorticoid-induced ON.
Our study also showed that early EPCs were impaired in
migration and secretory capacity but ECFCs had a significantly
impairment in proliferation and angiogenesis, which may reflect
different roles they played in the development of endothelial dys-
function in GC-induced ANFH. Early EPCs, also refer to angiogenic
monocytes, are the first cells to be recruited to sites of damage and
VEGF, HGF, G-CSF, IL-8 [17,33], which might create the ideal pro-
visional environment for genuine EPCs building vessels. And ECFCs
enhance neovasculogenesis by robust angiogenesis capacity and
provide sufficient number of endothelial cells based on their high
capacity of early EPCs and impaired proliferation and angiogenesis
potential of ECFCs may probably constitute an underlying mecha-
nism in endothelial dysfunction which takes place in the initiation
and progression of GC-induced ANFH.
Since EPCs could be detected relatively easily and safely, mea-
surement of the circulating EPCs number might be applied as a
valuable diagnostic test to identify ANFH. Besides, elevation of the
circulating EPCs number could be considered as a target of thera-
peutic interventions for improving endothelial dysfunction. Some
experimental studies of cell implantation using EPCs have been
undertaken to promote vascularization in patients with ischemic
disease. And recently, one study shows that implantation of EPCs
could increase vascularization and bone regeneration in the early
stages of ON of the femoral head in a rabbit model .
As an excellent cell source for vascular engineering strate-
gies, ECFCs have already been used in coimplantation strategies
with stabilizing cells or carrier-based delivery systems in human
skin substitutes . Additionally, Further studies have reported
the success in large-scale enrichment of ECFCs using a modified
bers for therapeutic applications and make ECFCs promising in
regenerative medicine and tissue engineering. And it could also be
C. Chen et al. / Joint Bone Spine 80 (2013) 70–76
used in GC-induced ANFH models as well, which would be a hot
topic in future study.
Disclosure of interest
The authors declare that they have no conflicts of interest con-
cerning this article.
This study was supported by the National Natural Science Foun-
dation of China (No. 30973044 and No. 81101375) and the Natural
Science Foundation of Hubei Province of China (No. 2009CDB046).
at the Department of orthopedic surgery of Union Hospital, Tongji
Medical College, Wuhan, the People’s Republic of China.
Appendix A. Supplementary data
Supplementary material (Table. S1) associated with this
article canbe found at http://www.sciencedirect.com,
 Saidi S, Magne D. Interleukin-33: a novel player in osteonecrosis of the femoral
head? Joint Bone Spine 2011;78:550–4.
 Séguin C, Kassis J, Busque L, et al. Non-traumatic necrosis of bone (osteonecro-
sis) is associated with endothelial cell activation but not thrombophilia.
Rheumatology (Oxford) 2008;47:1151–5.
 Kerachian MA, Harvey EJ, Cournoyer D, et al. Avascular necrosis of the femoral
head: vascular hypotheses. Endothelium 2006;13:237–44.
 Asahara T, Murohara T, Sullivan A, et al. Isolation of putative progenitor
endothelial cells for angiogenesis. Science 1997;275:964–7.
 Herbrig K, Haensel S, Oelschlaegel U, et al. Endothelial dysfunction in patients
with rheumatoid arthritis is associated with a reduced number and impaired
function of endothelial progenitor cells. Ann Rheum Dis 2006;65:157–63.
function, and cardiovascular risk. N Engl J Med 2003;348:593–600.
 Rabelink TJ, de Boer HC, van Zonneveld AJ. Endothelial activation and circu-
lating markers of endothelial activation in kidney disease. Nat Rev Nephrol
 Feng Y, Yang SH, Xiao BJ, et al. Decreased in the number and function of cir-
culation endothelial progenitor cells in patients with avascular necrosis of the
femoral head. Bone 2010;46:32–40.
 Lin Y, Weisdorf DJ, Solovey A, et al. Origins of circulating endothelial cells and
endothelial outgrowth from blood. J Clin Invest 2000;105:71–7.
 Ingram DA, Mead LE, Tanaka H, et al. Identification of a novel hierarchy of
endothelial progenitor cells using human peripheral and umbilical cord blood.
 Yoder MC, Mead LE, Prater D, et al. Redefining endothelial progenitor cells
via clonal analysis and hematopoietic stem/progenitor cell principals. Blood
 Bellik L, Ledda F, Parenti A. Morphological and phenotypical characterization
of human endothelial progenitor cells in an early stage of differentiation. FEBS
 Jacobs B. Epidemiology of traumatic and nontraumatic osteonecrosis. Clin
Orthop Relat Res 1978;130:51–67.
the lessons learned from a comparative study of osteonecrosis in man and
experimental animals. Vet Pathol 2003;40:345–54.
 Wu X, Yang S, Duan D, et al. Experimental osteonecrosis induced by a combi-
nation of low-dose lipopolysaccharide and high-dose methylprednisolone in
rabbits. Joint Bone Spine 2008;75:573–8.
 Wu X, Yang S, Duan D, et al. A combination of granulocyte colony-stimulating
factor and stem cell factor ameliorates steroid-associated osteonecrosis in rab-
bits. J Rheumatol 2008;35:2241–8.
 Hur J, Yoon CH, Kim HS, et al. Characterization of two types of endothelial
progenitor cells and their different contributions to neovasculogenesis. Arte-
rioscler Thromb Vasc Biol 2004;24:288–93.
vascularization and regeneration. Nat Med 2003;9:702–12.
cells from AC133-positive progenitor cells. Blood 2000;95:3106–12.
 Peichev M, Naiyer AJ, Pereira D, et al. Expression of VEGFR-2 and AC133 by cir-
culating human CD34(+) cells identifies a population of functional endothelial
precursors. Blood 2000;95:952–8.
cells are not derived from CD133+ cells or CD45+ hematopoietic precursors.
Arterioscler Thromb Vasc Biol 2007;27:1572–9.
 Yamamoto H, Kato H, Uruma M, et al. Identification of two distinct populations
of endothelial progenitor cells differing in size and antigen expression from
human umbilical cord blood. Ann Hematol 2008;87:87–95.
 Rehman J, Li J, Orschell CM, et al. Peripheral blood “endothelial progenitor
cells” are derived from monocyte/macrophages and secrete angiogenic growth
factors. Circulation 2003;107:1164–9.
 Sieveking DP, Buckle A, Celermajer DS, et al. Strikingly different angiogenic
properties of endothelial progenitor cell subpopulations: insights from a novel
human angiogenesis assay. J Am Coll Cardiol 2008;51:660–8.
 Medina RJ, O’Neill CL, Sweeney M, et al. Molecular analysis of endothelial pro-
genitor cell (EPC) subtypes reveals two distinct cell populations with different
identities. BMC Med Genomics 2010;3:18.
 Rabelink TJ, de Boer HC, de Koning EJ, et al. Endothelial progenitor cells: more
lial cell precursors. Am J Physiol Heart Circ Physiol 2007;292:H1–18.
 Kerachian MA, Séguin C, Harvey EJ. Glucocorticoids in osteonecrosis of the
femoral head: a new understanding of the mechanisms of action. J Steroid
Biochem Mol Biol 2009;114:121–8.
 Vogt CJ, Schmid-Schönbein GW. Microvascular endothelial cell death and
rarefaction in the glucocorticoid-induced hypertensive rat. Microcirculation
 Wolff JE, Guerin C, Laterra J, et al. Dexamethasone reduces vascular den-
sity and plasminogen activator activity in 9L rat brain tumors. Brain Res
 Li X, Jin L, Cui Q, et al. Steroid effects on osteogenesis through mesenchymal
cell gene expression. Osteoporos Int 2005;16:101–8.
 Akaike M, Matsumoto T. Glucocorticoid-induced reduction in NO bioavailabil-
ity and vascular endothelial dysfunction. Clin Calcium 2007;17:864–70.
 Li Calzi S, Neu MB, Shaw LC, et al. EPCs and pathological angiogenesis: when
good cells go bad. Microvasc Res 2010;79:207–16.
 Sun Y, Feng Y, Zhang C, et al. Beneficial effect of autologous transplantation of
endothelial progenitor cells on steroid-induced femoral head osteonecrosis in
rabbits. Cell Transplant 2011;20:233–43.
 Reinisch A, Hofmann NA, Obenauf AC, et al. Humanized large-scale expanded
endothelial colony-forming cells function in vitro and in vivo. Blood
 Kolbe M, Dohle E, Katerla D, et al. Enrichment of outgrowth endothelial cells
in high and low colony-forming cultures from peripheral blood progenitors.
Tissue Eng Part C Methods 2010;16:877–86.